Optimizing Language Model's Reasoning Abilities with Weak Supervision
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866909192798863360 |
|---|---|
| author | Tong, Yongqi Wang, Sizhe Li, Dawei Wang, Yifan Han, Simeng Lin, Zi Huang, Chengsong Huang, Jiaxin Shang, Jingbo |
| author_facet | Tong, Yongqi Wang, Sizhe Li, Dawei Wang, Yifan Han, Simeng Lin, Zi Huang, Chengsong Huang, Jiaxin Shang, Jingbo |
| contents | While Large Language Models (LLMs) have demonstrated proficiency in handling complex queries, much of the past work has depended on extensively annotated datasets by human experts. However, this reliance on fully-supervised annotations poses scalability challenges, particularly as models and data requirements grow. To mitigate this, we explore the potential of enhancing LLMs' reasoning abilities with minimal human supervision. In this work, we introduce self-reinforcement, which begins with Supervised Fine-Tuning (SFT) of the model using a small collection of annotated questions. Then it iteratively improves LLMs by learning from the differences in responses from the SFT and unfinetuned models on unlabeled questions. Our approach provides an efficient approach without relying heavily on extensive human-annotated explanations. However, current reasoning benchmarks typically only include golden-reference answers or rationales. Therefore, we present \textsc{PuzzleBen}, a weakly supervised benchmark that comprises 25,147 complex questions, answers, and human-generated rationales across various domains, such as brainteasers, puzzles, riddles, parajumbles, and critical reasoning tasks. A unique aspect of our dataset is the inclusion of 10,000 unannotated questions, enabling us to explore utilizing fewer supersized data to boost LLMs' inference capabilities. Our experiments underscore the significance of \textsc{PuzzleBen}, as well as the effectiveness of our methodology as a promising direction in future endeavors. Our dataset and code will be published soon on \texttt{Anonymity Link}. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_04086 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Optimizing Language Model's Reasoning Abilities with Weak Supervision Tong, Yongqi Wang, Sizhe Li, Dawei Wang, Yifan Han, Simeng Lin, Zi Huang, Chengsong Huang, Jiaxin Shang, Jingbo Computation and Language While Large Language Models (LLMs) have demonstrated proficiency in handling complex queries, much of the past work has depended on extensively annotated datasets by human experts. However, this reliance on fully-supervised annotations poses scalability challenges, particularly as models and data requirements grow. To mitigate this, we explore the potential of enhancing LLMs' reasoning abilities with minimal human supervision. In this work, we introduce self-reinforcement, which begins with Supervised Fine-Tuning (SFT) of the model using a small collection of annotated questions. Then it iteratively improves LLMs by learning from the differences in responses from the SFT and unfinetuned models on unlabeled questions. Our approach provides an efficient approach without relying heavily on extensive human-annotated explanations. However, current reasoning benchmarks typically only include golden-reference answers or rationales. Therefore, we present \textsc{PuzzleBen}, a weakly supervised benchmark that comprises 25,147 complex questions, answers, and human-generated rationales across various domains, such as brainteasers, puzzles, riddles, parajumbles, and critical reasoning tasks. A unique aspect of our dataset is the inclusion of 10,000 unannotated questions, enabling us to explore utilizing fewer supersized data to boost LLMs' inference capabilities. Our experiments underscore the significance of \textsc{PuzzleBen}, as well as the effectiveness of our methodology as a promising direction in future endeavors. Our dataset and code will be published soon on \texttt{Anonymity Link}. |
| title | Optimizing Language Model's Reasoning Abilities with Weak Supervision |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2405.04086 |